CA1131658A - Plaster consistency reducer - Google Patents

Abstract

Abstract of the Disclosure An additive for reducing the amount of water necessary in forming a pourable aqueous calcined gypsum slurry, the additive comprising a high potassium salt of a condensation product of naphthalene and sulfonic acid having a molecular weight between about 300 and 3,000. Aqueous plaster slurries containing the additive are pourable, moldable and castable with much less excess water than ordinarily required, resulting in dried cast products of Increased strength and other physical property benefits.

Description

-~13~5~3 ) PLhSTE~ CONSISTENCY ~EDUCER
B _ ground of the Invention 1 Field of the Invention .
The present invention relates to the process of a setting calcined gypsum slurry, and is more particularly concerned with a composition for reducing the amount of water necessary in form-ing a pourable aqueous calcined gypsum slurry to use in making cast or molded products.

2. Prior Art Plaster ~i.e. calcined gypsum) has long been a large volume commercial article of commerce either as various dry plaster compositions to be mixed by the user with water in forming cast gypsum articles or in manufactured products such as gypsum wall-board and cast art objects and the like. Generally, molds oc cast gypsum articles are manufactured by dispersing calcined gypsum and additives in sufficient water to form a pourable slurry, casting the slurry into a mold of desired shape, allowing the slurry to set, and drying excess water from the set article.
The gypsum setting reaction involves the reaction of calcium sulfate hemihydrate and water to form calcium sulfate dihydrateO
The theoretical water required to convert the calcined gypsum to set gypsum dihydrate is only 18.7% by weight on a pure basis~
However, considerable excess water is required to fluidize the calcined gypsum and obtain proper flow of the gypsum slurry Ln the casting or molding operation. The amount of excess water depends primarily upon the type of plaster particle, whether beta hemi-hydrate or alpha hemihydrate, and other additives in the plaster composition. Additives conventionally used in minor a~ounts include accelerators, retarder8, fibrous reinforcements, and con-sistency reducers. Consistency reduclng agents are typifie~ by the lignosulfonates, gum arabic, modified starches and other cell-ulosic derivatives. Ordlnarily, plasters produced at different .~

..

~L3 3L~8 locations from different sources will also vary in their water requirements to form a pourable slurry. Thus, calcined gypsum compositLons that are predominantly or substan~ially beta heml-hydrate may have a normal consistency in a range of about 60-90 or more cc. This rel~tes to a water usage of 60-90 cubic centimetecs of water per 100 grams of the calcined gypsum in ordes to form a readily pourable and flowable gypsum slurry. Other calcined gyp-Sum compositions that predominantly contain alpha hemihydrate may have a normal consistency of 34-55 cc. for example.
2. ~io~ A~t ~ he use of cellulosic additives as consistency reducers in calcined gypsum plasters has been suggested as early as around 1900.
In the area of another type of inorganic cementitious mater-ial, Portland cement, various other materials have been suggested as additives to modify properties of the composLtion. For example U.S. 3,465,825 suggests the use of the mixed salts of lithium and sodium of the condensation products of mononaphthalene sulfonic acid and formaldehyde as fluid loss control agents in Portland cement compositions. Such compositions are highly alkaline, gen-erally pH 11 and higher, and the use of such additives have been thought to be effective only in highly alkaline media. For cer-tain specific industrial applications it has been possible to combine substantial amounts of Portland cement compositions and plaster compositions. See for example U.S. 3,582,376; 3,847,635 and 3,852,081. Such compositions because of their high alkalinity due to the substantial quantlties of the Portland cementitious ingredients have taught the utilization of naphthalene sulfonLc acid and formaldehyde condensation produc~ consistency reducing additives. However, the available materials of this type have not been entirely effective or satlsfactory in non-highly alkaline ' 2 ~ 3~6~8 media and their usage has not been recommend2d in plaster composi-tlons whLch have not been adjusted to high alkalinity. In addition attempts to utilize commercially available condensates of thls type Ln plaster compositions has revealed a severe detrLmental characteristic of the condensate. Compositions containing sub-stantial amounts Oe Portland cements are generally a dark grey ln color; however, predominantly plaster containing compositions, especially those most usable in the manufacture of pottery and artware objects are very light in color and present a quite pleas-ing appearance. The addition of commercially available sodium form of naphthalene sulfonic acid and formaldehyde condensates resulted in a brown discoloration developing in the cast article on exposure to sunlight and with time, which discoloration i9 highly objectionable from a marketing standpoint.
_MMARY OF T~IE INVENTION
It has now been found that a highly effective plaster consis-tency reducing additive whLch does not cause objectionable cast discoloration may be prepared from commercially available naphtha-lene sulfonic acid and condensation products by controlled treat-ment with a basic compound containing potassium such as potassium hydroxide or potassium sulfateO
~ ccordingly, it is a principal object and advantage of the present invention to provide a plaster consistency reducer which is highly effective in decreasing the amount of water necessary to form a pourable aqueous calcined gypsum slurry on mixing wLth watec and which is not affected by exposure of the set composl-tion to sunlight and which does not require a highly alkaline media for its effectLveness.
It is a further object of the lnventLon to control the amount of water necessary to fosm a pourable slurry of a calcined gypsum composition. It is still another object of the Invention to ptQ-- '' '' ;

~L3~

vide a method for preparing cast gypsum articles using a less amount of water than normally required and to provide a product prepared by the said method having excellent strength and other physical properties including resistance to discoloratlon on exposure to sunlight.
According to the invention, a composition for reducing the water necessary to form a pourable slurry with powdered plaster compositions is prepared by partially neutralizing a commercially available condensation product oÇ naphthalene sulfonic acid with potassium, there being present at least about .lO~ by weight and pre~erably about 20-30~ by weight potassium as R2O based on the total weight of the condensation product. Generally from about l/2 to about 20 pounds of such additLve per ton ~2,000 pounds) of dry plaster composition may be used ln various difEerent calcined gypsum compositions, although a range of about S to about 15 pounds per ton is preferred.
DESCRIPT ~
The naphthalene sulfonic acid and formaldehyde condensation products, or condensed naphthalene sulEonates as they are some-times known in the trade, preferred in the present invention are anionc, polymer type dispersing agents supplied as light-colored powders or aqueous solutions having molecular weights varying between 300 and 3,000. They are prepared by condensing an alky-lene or napthalene and an aldehyde such as formaldehyde with sulfonic acids derived from various petroleum cracking processes to yield water and the acid form which is generally neutralized or partly neutralized with sodium hydroxide to give the sodium con-densate salt as an article of commerce. The organic structure of the materials has not been completely determined, may be somewhat variable, and does not appear to effect or concern the novel additives contemplated herein. Condensed naphthalene sulfonates , , 1~3~S~

suitable for use in the present inventlon are marketed by a number of companies under various trade names. A number of naphthalene sulfonic acId and Eormaldehyde condensate mater,ials are manufac-tured and marketed by Diamond Shanrock under the trademark "LOMAR"
and their lLterature lists chelnical analysis and physical charac-terLstics for a broad range of molecular sizes in the acid form or the sodium salt form. Other condensed naphthalene sulfonates suitable for use in forlaing the additive oE the present invention are produced by W. R. Gtace and Company under the trademark "DAXAD~ and their analysis and properties are produced in their technical literature.
The additives of the present invention are conveniently formed generally by partly neutralizing any of the available con-densed naphthalene sulfonates with a basic material containing potassium such as potassium hydroxide. It does not appear neces-sary to effectuate complete neutralization or replacement of other salt forms with the potassium forms; and it has been found comple-tely satisfactory to merely provide for the presence of potassium in an amount from about .l0 to 30~ by weight of the condensate.
Such may be accomplished by any convenient means such as dissolv-ing potassium hydroxide or potassium sulfate into the liquid acid form of the condensation product and drying and pulverizing the materials by any suitable means such as spray drying or conven-tional drying and grinding to form pulverulent materials or such means such as solubilizing the available powdered salt condensed naphthalene sulfonate such a_ the sodium form with aqueous potas- -sium hydroxide followed by drying and grinding of the mixture of potassium and condensate mixtures.
DESCRIPTION OF SPECIFIC EMBODIMENTS
The plaster compositions may be based upon any conventional calcined gypsum material including alpha calcium sulfate hemi~

S

.r .
., , ~ . , -~:3L3~L~S~

hydrate and beta calcium sulfate hemihydrate formed by varLous procedures. Such materials may be derived by calcination oE
natural gypsum rock or as a by-product of various chemlcal pro-eesses. The calcined gypsum compositions may contain typical additives conventionally used in mlnor amounts to modify proper-ties hereof for preferred commercial products.
The manner and order of incorporating the potassium form naphthalene sulfonate condensates of the invention with the cal-cined gypsum is not critical. Thus the condensate may be mixed with the dry calcined gypsum or the already formed aqueous slurry of calcined gypsum. The condensate may be added e;ther separately or together with conventional additives to the calcined gypsum ln forming dry packaged industrial plaster compositions, or metered ~n powdered or water solubilized form into the aqueous slurry sep-arately from or in combination with conventional additives durlng gypsum casting or molding operations.
In order to meet consistency and physical property require-ments, particularly high density, high strength, and high su~faee hardness requirements necessary to the usage requirements in in-dustrial molding, pottery and art plaster industries, many of the plaster compositions for industrial casting are formulated with blends of beta caleium sulfate hemihydrate and alpha calcium sul-fate hemihydrate, commonly referred to as "alpha gypsum". ~l~ha gypsum is known for its lower water requirements and resultallt set cast higher density, higher strength, and higher surface hardness set cast. ~lowever, it is also known for its scarcity in produe-tion and commands a premium price because of special production requirements necessary to obtain this form of caleined gypsum.
In a series of evaluations, special industrial casting , plaster composition formulations were modified to substltute small ! amounts of the water reducing agent of the present Inventlon or an .
!
- ., -- .

3L~31~

unmodifLed condensed naphthalene formaldehyde sulfonate into commerclal products or the plaster bases for such products then evaluated with respect to physlcal properties; water absorptlon and cast discoloration.

In a first series of evaluations the commercially available LOMAR D sodium salt form of condensed naphthalene sulfonate was treated to partially replace the sodium with potassium in deter-mining consistency reducing efficiency in a plaster composition of alpha calcined gypsum without the usual additives for commer-cial plaster formulations.
To incorporate the various amounts of potassium salt, 50 yram aliquots of the commercial condensate LOMAR D, sodium salt form of naphthalene sulfonic acid and formaldehyde condensate which is a powder containing sodium equivalents of up to 11~ Na2SO4, were dissolved in 200 cc. oE deionized water. The water was at room temperature, and the condensate was completely solubilized in the water. Varying amounts of potassium hydroxide or potassium sulfate were then added to the water solubilized condensate and thoroughly blended in a high speed mixer for 15 minutes. The resultin~ mixture was then dried at 250 F. to dryness and the resulting powder was ground to pass a 50 mesh lU.S. Standard) sieve. rrhe thoroughly dried and crushed mixture was added in various amounts to an alpha calcined gypsum plaster base and evaluated for reduced water requirements (measured as normal con~
sistency to achieve a pourable slurry) and setting time by stan-dard hand set 300 gram Vicat set (ASTM C-472).

~3~L~5~3 Table I

Add ition Normal Setting Formulation Lcvel~ ConsistencyTime Control A -Alpha Gypsum -- 35-36 1/2 20-30 Alpha Gypsum with Potassium Add it ion at:
#5 - 5% ROH 5 26 17 #6 _ n 10 19.5 14 #7 - " 15 19 20 #8 - 1% KOH S 27.5 18 #9 - " 10 21 13 #10 - " 15 19 13 #11 -0.5% KOH 5 27.5 13 tl2 - " 10 22 13 ~13 - " 15 19 10.5 tl4 -0.25~ KOH 5 27.5 15 i~15 -- - 10 22 13 #16 - " lS 19 11 #17 - 5% K2SO4 5 27.5 15 - #18 - " 10 20.5 10 #19 _ n 15 18 9 # 20 - 1 % K2SO4 5 27.5 13.5 #21 - " 10 ' 21.5 10 #22 _ n 15 18.5 9 # 23 -0.25% K2SO4 5 27.5 13 #24 " 10 21.5 8 #25 n 15 18.5 7.5 * lbs./ton (2,000 lbs.) ' ' , ' ' .

,

3~3 6S~3 Typical results are set forth in Table I, reporting varlous amounts by weight of potassium hydroxidc or potassium sulfate added to the aqueous mixture. From Table I it ls quite clea~ that very little potassium augmentatlon need be provided to the sodlum salt form of condensate in order to achleve very substantial re-ductions in water requirements. Quite surprisingly, at each usage level of the additive of the invention as set forth in Table I
substantially the same water reduct.ion effect was obtained regard-less of the amount of potassium provided for replacement of sodlum salt in the condensate.
EXAMPLE _ In anotl-er series of evaluations a high potassium salt form of the condensate of the invention was compared to the high sodlum salt form of condensed naphthalene sulonate using commercially available forms of LOM~R D, the one form containing 22.5~ potas-sium as K20 and the other containing 21% sodium as Na20. They were evaluated in a low consistency high strength commercial cast-lng plaster containing alpha gypsumO
For the first evaluation the commercial plaster composition, Control B, was augmented with those amounts of the condensate salt addition as would provide almost the same consistency reduction and evaluated for physical properties, as set forth in Table II.

.

~3~6~

T~LE II

Commercial Plus Condensate Alpha Gypsum Com~osition Sodium Potassium (Control B) AdditIve --- 11 lbs./ton 5 lbs./ton Normal Consistency 35 cc. 27 cc. 28 cc.
Set Time 21 minutes 25 minutes 25 minutes Time of Temperature Rise After Set 2a minutes 34 minutes 36 minutes Dry Cast Compres-sive Strength 7579 p.s.i. 11992 p.s.i. 12500 p.s.i.

Dry Cast Density lbs./cubic foot 102.3 117.~ 116.0 Two Week Slab Discoloration Wet None Considerable Very Slight Dry None Considerable Very Slight From Table II it can be seen that in order to achieve the same consistency results with the sodium form, the addition level had to be increased to 11 pounds per ton compared to 5 pounds per ton for the potassium form. This is a substantial jump in usage requirements in order to obtain substantially equivalent perfor-mance.
In order to meet usage requirements in the industrial mold-ing, pottery and art casting industry, plaster compositions must not only provide high compressive strength but also must be resis~

tant to sunlight discoloration, staining, and water absorption.
To evaluate for discoloration, the alpha calcined gypsum plaster base of the industrial casting plaster composition was evaluated by adding varying amounts of the condensates to the plaster base and mixing witll water. The slurry was then formed into rectangular slabs measuring 3 inches by 14 in~hes by 3/8 inch and exposed to sunlight for a two wcek period of time. Some of lo .

~3~s8 the samples ~ere placed immediately after demolding on a wLndow ledge that received sunlight all day and reported as "wet" results while duplicate samples were oven dried to constant weight before placing In sunlight and reported as "dry" results. Results are set forth in Table II as "two week slab discolorationN results. -Thc sodium form resulted in discoloration of the cast that would be commercially objectionable while the potasSium form condensate of the invention did not.
In another evaluation in tl)is series, varying amounts of the condensate were added to the plaster base of the gypsum composi-tion and evaluated for water consistency reduction as set forth in Table III.
TABLE III

Consistency Change From Formulation~sage Level Consiste~ Control __ Control C
(Plaster base of Control B) --- 35 cc. ---Sodium Salt Condensate ~21% Na2O)5 lb./ton 31 cc. -11~
10 lb./ton 29 cc. -17%
15 lb./ton 27 cc. -22%

Potassium Salt Condensate (22.5% R2O)5 lb./ton 31 cc. -11%
10 lb./ton 23 cc. -34~
15 lb./ton 20 cc. -42%

It may be clearly seen from Table III that the potassium con-densate of the invention additive was much mo~e effectlve in reducing water requirements at the same levels of usage as the sodium salt condensate. Each of the 10 and 15 pound per ton addl-tion levels of the potassium salt condensate effectuated twlce as much change in consistency over the control as did the comparable sodium salt condensate. Further~ it may be extrapolated from .

1~31ti,51~ r Tables II and IV that the sodium salt form at the higher usage level would be commeccially objectionable for providing discolora-tion of cast articles.
In another evaluation in thls series two different high strength low consistency industrial casting plaster formulations containing alpha gypsum were compared to a plaster of Paris plaster composition containing customary additives and furthee -:
including either the preferred potassium condensate additive of the invention or its most comparable sodium salt form condensate, LOMAR D condensate and the samples evaluated for physical proper-ties as set forth in Table IV.

T~BLE IV
Formulation Control Control ients D 1 2 E 3 4 Commercial Industrial Plaster 100~ -- 100% --- ---Plaster Composition --- 99.92% 99.92% -- 99.94% 99.94 Condensate -Sodium Salt assay) . .08% --_ __ .06% -__ Condensate -Potassium Salt (22% K2O .08~ --- ___ .06%
Properties Normal Consistency cc 59 59 5~ 62 60 60 Set Time -Minutes 22 22 19 21 17 16 Temp. Risel 17 17 19 25 lt 18 Dry Cast Strength psi 2,998 2,802 2,723 2,888 2,820 2,762 t, - ~
. .. .

: : :

:

~3~L~58 .
TA~LE IV (Continued) ~ormulation Control Control Ingredients D 1 2 E 3 4 PrOPQr ties Dry Cast Density lb/cu ft 72.5 72.2 71.6 72.9 72.7 72.3 Setting ExpallsLon 0.174~ 0.187~ 0.216% 0.238% 0.196% 0.221%
Ring Test2 Water Absorption cc/min. 2.75 4.90 2.6S
Weight Gain3 at 5 min. 22.5% 24.1% 21.3%
10 min. 23.~% 26.1% 22~9~
15 min. 24.1% 27.4% 24.3%
30 min. 25.4% 28.7% 25.6%
1 hour 26.4% 29.1% 26.3%
Wet4Out/Dry-Out % wt. loss 3.9~ 3.8~ 4.0%
Control D 1 2 Two Week Slab Discoloration Wet None Considerable Slight Dry None Considerable Very slight 1 Temperature rise at: the time in minutes to reach the maximum temperature during the exothermic hydration of calcined gypsum setting to hydrated gypsum is indicative of the point in the reaction at which almost all of the plaster has rehydrated and the sample has obtained its maximum wet strength. Determined by casting a sample to a breadth of 2 inches in a container as described in ASTM C-472 and portionate with the thermocouple in an insulated cabinet then measuring the time during which the temperature rises.
Ring test: simulates water absorption characteristics to yive an indication of dewatering rate. The plaster is cast into a 2 inch cube shape, oven dried to constant weight and cooled to room temperature in a sealed bag to prevent moisture pick-up.
Then placed on a flat surface within a 1 inch inslde diameter by 1 inch high aluminum ring secured to the surface with a rope or modeling clay. Water (10 cc.) is poured into the ring and timed for absorption into the cast sample. If leakage under the ring is observed, the test is discontinued. The absorption rate into the sample is obtained by dividlng 13 cc. by the elapsed tlme for the water to absorb into the cast sample anc`i the value reported is the average of duplicate evaluations.

!

' ~ 1 .
~ 3 .

.

5~

Weight gain: Compares relative absorption rates of samples when 2 Inch cubes of the sample materials are fully emersed in water ~or I hour. Prior to emersion, samples are dried to con-stant weight, cooled to room temperature in a sealed plastlc bag, then weighed to a I gram accuracy. Thereaiter the samples are placed in an appropriate sized container to be completely covered with water throughout the I hour period of time. At S minute intervals, the sanlple is removed from the water, lightly blotted with a towel and weighed to the nearest gram, then returned to the water. Absorption rate as percentally gained is determined by subtracting the initial dry weight from the wet weight at the timed interval, dividing by the initial dry weight and multiplying by 100~

4 Wet-out dry-out tests were conducted on oven dried 2 inch cubes by weighing the dried cubes, Eully emersing them in water for 4 hours, reweighing them, then drying the samples for 20 hours in a 110~ F. oven and reweighing again. This procedure is contin-ued Eor a 2-week period to determine if the formulations eroded at comparable ratcs to simulate manner in which they may erode in pottery shop usage.
From Table IV it may be clearly seen that the potasslum additive provided greater resistance to discoloration and less water ab-sorption in plaster compositions for industrial casting while maintaining other physical properties at acceptable levels.

As set forth in Tables III and IV above additives may be included in their conventional amounts for their usual purposes, such as accelerators, retarders, defoamers and the like. Other ~ adjuvants in mo~e major amounts may also be added without dellter-ious affects in the use of the additives of this invention. For example, Portland cement may be added in varying amounts without deleterious affects. In fact there appears to be an enhanced con~
sistency efficiency with the condensate of the invention, at least at low levels of Portland cement augmentation.to calcined gypsum plas'er of Paris compositions, as set forth in Table VO

. .

- -: -' ' ~

6~8 TABLE Y
____ Alpha Gypsum Additive ~ddition Level Formulation: Plus 5~ Port-land CementSodIu~ Potass _m tControl E r11 lb ~ton 5 lb~ ton PropertLcs Normal Consis- .
tency ~cc.) .40 27 28 SettIng Time (Minutes) 20 25 25 Time of maximum Temperature Rise ~fter Set (minutes) 20 26 19 Dry Density (lbs./cubic ft.) -- 123 125.7 Dry Compressive Strength 6125 12,217 13,866 Set Expansion 0.250~ .255~ . .271 Two Week Slab Discoloratlon Wet None Considerable None Dry None Considerable None .' ' '~' '' . ' ' ' , ' .

Claims (17)

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A plaster consistency reducing agent for reducing the amount of water required to be mixed with a dry plaster composition to form a pourable aqueous slurry consisting essentially of a potassium salt of naphthalene sulfonic acid condensate having a molecular weight between about 300 and 3,000, the potassium being present in an amount of about .10% to about 30% by weight expressed as K2O, by weight of the condensate.

2. The plaster consistency reducing agent of Claim 1 containing about 10% to about 30% potassium expressed as K2O.

3. The plaster consistency reducing agent of Claim 1 in which the condensate is naphthalene sulfonic acid formaldehyde condensate and contains about .10% to about 30% by weight of the condensate of potassium expressed as K2O.

4. The plaster consistency reducing agent of Claim 1 comprising the product of solubilizing a naphthalene sulfonic acid condensate in water with about 0.1% to about 5% by weight of a basic potassium compound selected from the group consisting essentially of potassium sulfate and potassium hydroxide;
drying the solubilized material to a solid; and grinding the solid to a powder.

5. A plaster composition consisting essentially of calcined gypsum and a minor amount of a water-reducing agent for the calcined gypsum, the water-reducing agent comprising a potassium salt of naphthalene sulfonic acid condensate having a molecular weight between about 300 and 3,000, said condensate containing from about .10% to about 30% by weight of potassium expressed as potassium oxide.

6. The plaster composition of Claim 5 in which the condensate is naphthalene sulfonic acid formaldehyde condensate and contains about .10%
to about 30% by weight of the condensate of potassium expressed as K2O.

7. The plaster composition of Claim 5 wherein the calcined gypsum is alpha calcined gypsum calcium sulfate hemihydrate.

8. The plaster composition of Claim 5 wherein the calcined gypsum is beta calcium sulfate hemihydrate.

9. The plaster composition of Claim 5 wherein the water-reducing agent is present in the plaster composition in an amount of from about 1/2 pound to about 20 pounds per ton.

10. The plaster composition of Claim 5 wherein the water-reducing agent is present in the plaster composition in an amount of about 5 pounds to about 15 pounds per ton.

11. The plaster composition of Claim 5 wherein the potassium salt contains about 20% to about 25% potassium expressed as potas-sium oxide.

12. The plaster composition of Claim 5 wherein the conden-sate is naphthalene sulfonic acid formaldehyde condensate contain-ing about 10% to about 30% by weight of potassium expressed as K2O and the condensate is present in the plaster composition in an amount of from about 5 pounds to about 15 pounds per ton.

13. A method of reducing the amount of water required to be mixed with a dry plaster composition to obtain a pourable slurry which comprises adding to a calcined gypsum plaster composition from about 1/2 pound to about 30 pounds per ton of naphthalene sulfonic acid condensate having a molecular weight from about 300 to about 3,000 and containing by weight of the condensate from about 10% to about 30% of potassium expressed as potassium oxide.

14. The method of Claim 13 wherein the condensate is added to a dry plaster composition.

15. The method of Claim 13 wherein the condensate is added to an aqueous slurry of calcined gypsum.

16. The method of Claim 13 wherein the condensate contains about 20 to about 25% by weight of potassium expressed as potas-sium oxide, and the condensate is added in an amount of about 5 to about 15 pounds per ton of dry calcined gypsum plaster.

17. The method of Claim 13 wherein the condensate is naph-thalene sulfonic acid formaldehyde condensate containing about 10%
to about 30% by weight of potassium expressed as K2O and the condensate is added in an amount of about 5 pounds to about 15 pounds per ton.